Mythical Man-months, Microservice Architectures and a Million Developers

Leave a reply

farm-workersWatching a presentation on scaling microservices by Adrian Cockcroft http://perfcap.blogspot.com/  Adrian mentioned Twitter had over 1M developers using their API's. DEVELOPERS, not API calls or Apps using the API's etc. WOW!

So when Fred Brooks published "The Mythical Man-Month" in 1975 (MUST reading for all managers esp tech) he described solving the people scaling problem at IBM using clever team designs. His "Perfectly Partition-able Task" (like picking corn or strawberries) stuck in my head (I read it in the early 80's) as a framework ever since.

The gist was that teams working on complex systems had a communications/coming up to speed burden proportional to (N^2-N)/2 , where N is the number of team members (developers). So if we can scale "Dev Ops" unbounded using these architectures, I believe we have a tangibly different world than the words "big data", "cloud computing" or scalable software captures.

Partitioning tasks changed the size of things we built in the physical world... pyramids, Ford's assembly line, etc. it will do the same for what we build in the information world. A much bigger deal than many recognize...

Paul Tudor Jones: 4 Ways to Make Money

8 Replies

At the 25th anniversary of the crude oil contract on the NYMEX, Paul Tudor Jones ( http://en.wikipedia.org/wiki/Paul_Tudor_Jones ) gave a presentation where he described how he categorizes trading strategies. Here they are with some examples...

1) Specialized knowledge of an Instrument

Stock picking long/short (research analysts)
Floor traders (back in the day)
Insider trading

2) Arbitrage (take advantage of pricing inefficiencies)

NYC vs. London, Cash vs. Futures
Pairs trading (statistical arbitrage or relative value)
Microstructure (Bid/Ask spreads) (HFT)

3) Trend (Momentum) Trading

Trin trading (day trading)
Nasdaq late 90’s
CTA “Trend-followers”

4) Execute at the “Peak of Human Emotions”

Small group of people (Idiot Savants) Exuberance vs Fear
Buffett “Be greedy when others are fearful and fearful when others are greedy”

This is a useful framework...if it's good enough for PTJ it works for me.

folders

Diversification 101 - why 8 is enough

8 Replies

diversified eggs

Harry Markowitz  ( http://rady.ucsd.edu/faculty/directory/markowitz/ ) is credited with the comment that diversification is the only free lunch in investing.

Mathematically diversification is about UNCORRELATED movement of asset prices. There are many ways people tackle this including MPT,  Post-MPT,  risk-parity and others.   I focus on the number .707. (I used it when designing microprocessor chips worrying about how many electrons I needed to get in place on time)

It is a VERY useful rule of thumb when combining assets with random returns (iid) that have LOW CORRELATION. Each time you DOUBLE the basket of assets the variance (standard deviation of returns) will be reduced by .707.  For example...

  • 1 Asset  volatility  N
  • 2 Assets volatility ~.707 * N   or   30% less
  • 4 Assets volatility ~.707^2 * N  or 50% less
  • 8 Assets volatility  ~707^3 * N  or 65% less
  • 16 Assets volatility ~.707^4 * N or 75% less
  • etc.

Notice the reduced bang for the buck. The reality is that it is VERY difficult to find 8 uncorrelated assets and getting harder with globalization (global equity markets have become more correlated the last 15 years). The biggies are Cash, Equities, Bonds, Real Estate, Absolute Return, Managed Commodities and Gold.

I don't like it when large  funds claim higher diversification because they have small positions. This is usually driven by liquidity constraints rather than diversification optimization and pushes returns to the mean.  Yes a bankrupt company is a RISK FACTOR that gets diluted with large positions but equities are equities. Buffett has said 10 and Cramer often works around the 8-10 range.

So focus on getting close to 8.  Nine won't help that much and the leap to 16 is tough even for Bridgewater Associates ( http://www.bwater.com ) and David Swensen ( http://en.wikipedia.org/wiki/David_F._Swensen ) did well by adding just one (timberland) to the Yale endowment.

Business Development title is up for grabs

Leave a reply

There's been lots of activity on LinkedIn groups discussing the role of Business Development or BizDev.  I found it interesting since I've occupied this slot and hired/promoted people into the slot.

This is a title that is being used to describe many roles now, including:

  1. Inside Sales
  2. Territory Sales
  3. All or part of Marketing Product Management
  4. Social Media content management

and others...

What surprises me is people in the first decade of their career with the title. I am schooled that BizDev is bolted to the CEO or division level executive leadership and working 3-10 years out. Self-driving cars, contact lenses and fiber to the home is BizDev for Google.

What are you going to do when your market saturates? All products, markets and industries mature and create challenges for the leadership. I believe the role should be led by people after accomplishments in Marketing/Sales and General Management with solid foundations in finance and some law. They should be able to sell stakeholders on how taking an "out of bounds" risk makes sense and be able to manage an acquisition or divestiture.

It is sometimes a convenient title for  founders if the company grows up around them. This is fine if they grow into the role. Don't mistake a practical people decision with the strategic role longer term.

Worst case I see the title sometimes being used to avoid using "sales" in someones title, the dead giveaway here is when the title is assigned with  1 year sales metrics.

Sloppy strategic thinking...

iPhone 4s was the Vax 11/780 of Pocket Computers

1 Reply

Old engineers remember the Digital Equipment Corporations' VAX 11/780 computer. The Vax 11/780 was the "benchmark" unit of computing (1 MIP) in the engineering world of the 1980's. Competitors had to compare their performance to the VAX 11/780. It was the machine Unix grew up on.   http://en.wikipedia.org/wiki/VAX-11

I think the iPhone 4s  became the first "benchmark" of pocket computers  (i.e. smartphones). It hit the "sweet spot" of cost/performance running a general purpose dual core iOS 5.0 (shared with iPods) with a MAC look and feel and properly virtualized i/o.  http://en.wikipedia.org/wiki/IPhone_4S

iphoneVax

Apple's Down Market Strategy

3 Replies

fishjumpApple is way out-thinking Wall Street as they move down market.  "Talking-heads"  at Bloomberg, NBC, and securities analysts have argued they need  cheaper phones to stem the market share growth of Android. Apple has kept it's cool and devised the best of all compromising strategies... BOTH.

Both a) more affordable phones and b) maintaining revenue and margins for the premium product/ecosystem.  They did this by taking a strategy lesson from BMW and Mercedes. Enter lower price markets by selling used phones at Android pricing, and replace the high-end  with enhanced functionality (i.e. AC7). Would you rather have a used BMW 3 series or a Ford Focus?

Steve Jobs long ago recognized how to use Moore's law to fill out product portfolios. Rather than designing high, medium and low end products, he would ride Moore's law cost reductions to sell yesterday's functionality at lower prices. The Apple software and manufacturing ecosystem rides along for free. Once customer loyalty is captured, it is easier to up-market sell the high end. Engineered cost-reductions can focus on the most bang for the buck products.

Now Apple's marketing strategists have convinced the organization to recycle 4, 4s and 5 phones by providing trade-ins.  It extracts the residual value of Apple phones into Apple's pocket, helps the high and low-end customer, and is ecological to boot.

This is an example of a key idea in good strategy. Good strategy increases efficiency somewhere in the value chain. In Apple's case it is extracting value from the oligopolistic carrier distribution channels by leveraging it's stores. Retail became a necessary evil when telcos had to sell phones to people (mobile) instead of buildings (landline).  Apple uses retail to build customer relationships (genius bars) not an every 2-year visit to re-new a contract.

Apple knows these machines are "pocket computers" first and phones second 🙂

Asymptotic Thinking

Leave a reply

asymptote

What if... a single machine could provide all of the goods and services that a country's economy currently produces?   Would human productivity (GDP per labor hour) be infinity ? Who should own the machine? Google? The government? Goldman Sachs? Would the end state look the same if we built such a machine over 100, 50, 10 or 2 years?

This is not rambling, this is "asymptotic thinking" and I believe it is a pre-requisite to good strategic thinking. In calculus it is  

   where p=human productivity and GDP=Gross Domestic Product.

This thinking is important because we approach limits incrementally from a DIRECTION and in strategy, as well as calculus, the direction matters.  More specifically there can be discontinuities along the path, much like crossing a chasm in product marketing.

How many strategies are there?

Leave a reply

Twenty?  As many as there are companies?

At the highest level of competitive differentiation there are 3 (but it is still hard to get it right).

Technology changes...strategy doesn't.  Change happens faster....strategy doesn't.

THE WORLD'S MOST FAMOUS BUSINESS-SCHOOL PROFESSOR IS FED UP WITH CEOS WHO CLAIM THAT THE WORLD CHANGES TOO FAST FOR THEIR COMPANIES TO HAVE A LONG-TERM STRATEGY. IF YOU WANT TO MAKE A DIFFERENCE AS A LEADER, YOU'VE GOT TO MAKE TIME FOR STRATEGY.

http://www.fastcompany.com/42485/michael-porters-big-ideas

PorterStrategies

 

"Plan" vs "Strategic Plan"

Leave a reply

PlanWhat's the difference between a plan and a strategic plan? Most companies, departments and people have a plan. So what's unique about adding the word "Strategic". It is not just an adjective, added to appear more sophisticated or smarter. It should change the content and mold objectives as well as articulate action steps.

Here's why:

1) Critical Environmental variables WILL change over the "planning horizon"

2) Actual outcomes will MOST LIKELY be different (sometimes  a lot, sometimes a little)

3) It should FORBID certain resource allocations to enhance the CONCENTRATION of them

4) Plans are NEVER FULLY ACHIEVED  but assume a next interval of time sequence

A "Plan" can assume the critical environment it is operating in will remain relatively stable over the "planning horizon". A "Strategic Plan" cannot and should not. Liddell Hart in his military strategy classic has a chapter on "The Concentrated Essence of Strategy and Tactics".  Out of 8 items he includes "Take a line of operation which offers alternative objectives" which is rarely understood among business strategic planners.